Inactivation of the NF1 gene is the underlying cause for one of the most common genetic diseases of the nervous system, neurofibromatosis type 1. Neurofibromin, the gene product of NF1 encodes for a 3000 amino acid protein that contains a rasGTPase activating domain (rasGAP), a negative regulator of the ras pathway. To date, prevailing evidence indicates that all pathologies associated with neurofibromatosis are the consequence of deregulation of ras signaling. The most common tumor pathologies in afflicted individuals are neurofibromas (100% incidence), malignant peripheral nerve sheath tumors (15% incidence), optic gliomas (pilocytic astrocytomas;20% incidence), and astrocytomas (<1% incidence). We have attempted to model these tumors in mice as a means of understanding the origin, molecular details of progression, and to develop potential therapies. Initial development of mouse knockout models yielded considerable information on NF1 in development, the null phenotype resulted in embryonic lethality thus limiting our ability to examine NF1 as a tumor suppressor (Brannan et al., 1994). The genetic trick of combining germline mutations at the NF1 and p53 tumor suppressors resulted in mice that developed MPNSTs with 100% penetrance. This result provided the first indication that mutations in the mouse NF1 genes could mimic the human tumor condition effectively (Vogel et al., 1998). To achieve better control of the NF1 modeling, we have developed tissue specific mutations by use of cre/lox technology (Zhu et al., 2000). Using conditional knockouts, we have successfully modeled formation of plexiform neurofibromas, thus identifying the local cell of tumor origin and the importance of a distant partner in tumor development. Much of this application focuses on our conviction that the distant partner is a mast cell. In addition we have new models that exhibit tumors of the CNS: namely optic tract gliomas and astrocytomas. In the present application we propose to develop a greater understanding of the genesis of optic gliomas. In addition, we will interface with our colleagues Dr. Clapp and Dr. Ingram in molecular studies of mast cell contribution to neurofibroma formation